Test chained assignment detection for Python 3.14

Author: jorisvandenbossche

pandas/_libs/internals.pyx

@@ -1,6 +1,9 @@
 from collections import defaultdict
+import sys
+import warnings
 
 cimport cython
+from cpython cimport PY_VERSION_HEX
 from cpython.object cimport PyObject
 from cpython.pyport cimport PY_SSIZE_T_MAX
 from cpython.slice cimport PySlice_GetIndicesEx
@@ -20,6 +23,7 @@ from numpy cimport (
 cnp.import_array()
 
 from pandas._libs.algos import ensure_int64
+from pandas.errors import ChainedAssignmentError
 
 from pandas._libs.util cimport (
     is_array,
@@ -996,3 +1000,55 @@ cdef class BlockValuesRefs:
                 return self._has_reference_maybe_locked()
         ELSE:
             return self._has_reference_maybe_locked()
+
+
+cdef extern from "Python.h":
+    """
+    #if PY_VERSION_HEX < 0x030E0000
+    int __Pyx_PyUnstable_Object_IsUniqueReferencedTemporary(PyObject *ref)
+    {
+        return 0;
+    }
+    #else
+    #define __Pyx_PyUnstable_Object_IsUniqueReferencedTemporary \
+        PyUnstable_Object_IsUniqueReferencedTemporary
+    #endif
+    """
+    int PyUnstable_Object_IsUniqueReferencedTemporary\
+        "__Pyx_PyUnstable_Object_IsUniqueReferencedTemporary"(object o) except -1
+
+
+cdef inline bint _is_unique_referenced_temporary(object obj) except -1:
+    if PY_VERSION_HEX >= 0x030E0000:
+        return PyUnstable_Object_IsUniqueReferencedTemporary(obj)
+    else:
+        return sys.getrefcount(obj) == 2
+
+
+# # Python version compatibility for PyUnstable_Object_IsUniqueReferencedTemporary
+# IF PY_VERSION_HEX >= 0x030E0000:
+#     # Python 3.14+ has PyUnstable_Object_IsUniqueReferencedTemporary
+#     cdef inline bint _is_unique_referenced_temporary(object obj) except -1:
+#         return PyUnstable_Object_IsUniqueReferencedTemporary(obj)
+# ELSE:
+#     # Fallback for older Python versions using sys.getrefcount
+#     cdef inline bint _is_unique_referenced_temporary(object obj) except -1:
+#         # sys.getrefcount includes the reference from getrefcount itself
+#         # So if refcount is 2, it means only one external reference exists
+#         return sys.getrefcount(obj) == 2
+
+
+cdef class SetitemMixin:
+
+    def __setitem__(self, key, value):
+        cdef bint is_unique = _is_unique_referenced_temporary(self)
+        # print("Refcount self: ", sys.getrefcount(self))
+        # print("Is unique referenced temporary: ", is_unique)
+        if is_unique:
+            warnings.warn(
+                "A value is trying to be set on a copy of a DataFrame or Series "
+                "through chained assignment.",
+                ChainedAssignmentError,
+                stacklevel=1,
+                )
+        self._setitem(key, value)

pandas/_libs/tslib.pyx

@@ -267,6 +267,7 @@ cpdef array_to_datetime(
     str unit_for_numerics=None,
 ):
     """
+    TODO no longer up to date
     Converts a 1D array of date-like values to a numpy array of either:
         1) datetime64[ns] data
         2) datetime.datetime objects, if OutOfBoundsDatetime or TypeError

pandas/core/frame.py

@@ -47,6 +47,7 @@
     properties,
 )
 from pandas._libs.hashtable import duplicated
+from pandas._libs.internals import SetitemMixin
 from pandas._libs.lib import is_range_indexer
 from pandas.compat import PYPY
 from pandas.compat._constants import REF_COUNT
@@ -58,7 +59,6 @@
 )
 from pandas.errors.cow import (
     _chained_assignment_method_msg,
-    _chained_assignment_msg,
 )
 from pandas.util._decorators import (
     Appender,
@@ -511,7 +511,7 @@
 
 
 @set_module("pandas")
-class DataFrame(NDFrame, OpsMixin):
+class DataFrame(NDFrame, OpsMixin, SetitemMixin):
     """
     Two-dimensional, size-mutable, potentially heterogeneous tabular data.
 
@@ -4212,7 +4212,7 @@ def isetitem(self, loc, value) -> None:
         arraylike, refs = self._sanitize_column(value)
         self._iset_item_mgr(loc, arraylike, inplace=False, refs=refs)
 
-    def __setitem__(self, key, value) -> None:
+    def _setitem(self, key, value) -> None:
         """
         Set item(s) in DataFrame by key.
 
@@ -4296,11 +4296,11 @@ def __setitem__(self, key, value) -> None:
         z  3  50
         # Values for 'a' and 'b' are completely ignored!
         """
-        if not PYPY:
-            if sys.getrefcount(self) <= 3:
-                warnings.warn(
-                    _chained_assignment_msg, ChainedAssignmentError, stacklevel=2
-                )
+        # if not PYPY:
+        #     if sys.getrefcount(self) <= 3:
+        #         warnings.warn(
+        #             _chained_assignment_msg, ChainedAssignmentError, stacklevel=2
+        #         )
 
         key = com.apply_if_callable(key, self)
 

pandas/core/generic.py

@@ -305,7 +305,7 @@ def _from_mgr(cls, mgr: Manager, axes: list[Index]) -> Self:
         The axes must match mgr.axes, but are required for future-proofing
         in the event that axes are refactored out of the Manager objects.
         """
-        obj = cls.__new__(cls)
+        obj = object.__new__(cls)
         NDFrame.__init__(obj, mgr)
         return obj
 

pandas/core/series.py

@@ -32,6 +32,7 @@
     properties,
     reshape,
 )
+from pandas._libs.internals import SetitemMixin
 from pandas._libs.lib import is_range_indexer
 from pandas.compat import PYPY
 from pandas.compat._constants import REF_COUNT
@@ -43,7 +44,6 @@
 )
 from pandas.errors.cow import (
     _chained_assignment_method_msg,
-    _chained_assignment_msg,
 )
 from pandas.util._decorators import (
     Appender,
@@ -234,7 +234,7 @@
 # class "NDFrame")
 # definition in base class "NDFrame"
 @set_module("pandas")
-class Series(base.IndexOpsMixin, NDFrame):  # type: ignore[misc]
+class Series(base.IndexOpsMixin, NDFrame, SetitemMixin):  # type: ignore[misc]
     """
     One-dimensional ndarray with axis labels (including time series).
 
@@ -1058,12 +1058,12 @@ def _get_value(self, label, takeable: bool = False):
         else:
             return self.iloc[loc]
 
-    def __setitem__(self, key, value) -> None:
-        if not PYPY:
-            if sys.getrefcount(self) <= 3:
-                warnings.warn(
-                    _chained_assignment_msg, ChainedAssignmentError, stacklevel=2
-                )
+    def _setitem(self, key, value) -> None:
+        # if not PYPY:
+        #     if sys.getrefcount(self) <= 3:
+        #         warnings.warn(
+        #             _chained_assignment_msg, ChainedAssignmentError, stacklevel=2
+        #         )
 
         check_dict_or_set_indexers(key)
         key = com.apply_if_callable(key, self)